Choosing and using survey meters and contamination monitors

Two instruments, two different jobs, and a set of choices that decide whether either one is trusted. Detector types, what to write into a specification, how calibration and the daily source check work, and the survey technique that actually finds contamination.

Walk into most Bangladeshi nuclear medicine departments and you will find one radiation instrument, usually in a drawer, usually with a flat battery, usually last calibrated at some point nobody can name. It gets carried out when an inspection is expected. That is the practical position, and it exists because the two instruments a department needs do different jobs and neither one is any use until somebody trusts it.

The regulatory expectation itself is set out in radiation safety and regulatory compliance. What follows is the engineering underneath it: what to buy, how to keep it honest, and how to use it so that it finds something.

Two instruments, two jobs

A dose rate meter answers the question how much radiation is there here, and reports in microsieverts or millisieverts per hour. It is what you use to survey a room boundary, to check a delivery package, to decide whether a corridor is safe to stand in and to measure a patient before release.

A contamination monitor answers a different question: is there loose radioactive material on this surface. It reports counts per second, and it is sensitive rather than accurate. Its job is to find a spot of activity on a bench, a glove, a shoe or a floor tile that a dose rate meter would never notice.

Departments try to make one instrument do both, and it does neither well. A sensitive contamination probe saturates in the field beside a therapy patient. A dose rate meter sized for that field will not see a small technetium spill at all. Buy both, and buy the contamination monitor first if the budget only stretches to one, because in unsealed source work loose contamination is the failure that actually happens.

Detectors, and what each one is good for

The detector determines the instrument's character far more than the badge on the case does.

• Energy compensated Geiger Muller tube. The standard general purpose dose rate detector. Rugged, cheap, adequate over the diagnostic energy range once compensated. It responds slowly at low dose rates and it can saturate in a very high field, which matters more than most people realise.
• Ionisation chamber. Accurate across a wide energy range and well behaved at high dose rates, which is why it is the sensible choice for a department doing radioiodine or lutetium therapy. It reads slowly and it is less convenient to carry. Some types are vented and need a pressure and temperature correction.
• Sodium iodide scintillation probe. Very sensitive, so it finds small amounts of activity and is good for low level area survey and for locating a lost source. It over-responds at low photon energies and saturates quickly, so it is a finding instrument rather than a measuring one.
• Thin window pancake GM probe. The workhorse contamination probe. Large window area, responds to beta and gamma, and its geometry suits sweeping a bench. The mica window is fragile, and in a humid climate a damaged or unprotected window is the commonest reason a department's monitor stops working.
• Plastic scintillator hand and foot monitors. Fixed at the hot lab door. Worth it once staff numbers make individual hand monitoring at a bench monitor unreliable, which in practice means once the department runs more than one list a day.

The one specification detail that catches out even careful buyers is the dose rate range. If the department will ever handle a therapy activity or receive a generator, the instrument has to read up into the millisievert per hour range at the surface of a pot. A meter that tops out below that is not merely inconvenient, it is misleading, because some tube types fold back towards zero when they saturate. Approaching a source with an instrument that has jammed low is how people get doses they never find out about.

What to write into the specification

Instrument tenders in Bangladesh tend to be written around a model name, which either produces one bid or produces a substitute nobody evaluated. Specify the performance instead.

• The measured quantity and the units. Ambient dose equivalent rate for the survey meter. Counts per second for the contamination monitor, with any activity per unit area scale stated as being for a named radionuclide.
• The range and the overload behaviour. State the top of the range you need and require the instrument to indicate overload rather than fold back.
• Energy response. State the energies you actually work with, from the low energy of iodine 125 if you use it, through technetium at 140 keV, up to 511 keV if PET is anywhere in the plan.
• Probe window area and its protection. A large window finds contamination faster. A removable protective grid keeps it alive.
• Audio output. Not a luxury. A survey done by ear is done at the right speed and with eyes on the surface rather than on the meter.
• Battery type available in Dhaka. An instrument that needs an unusual cell is an instrument that will be flat when it matters.
• A check source supplied with the instrument, mounted or in a holder, so the daily response check is repeatable in the same geometry.
• Calibration certificate at delivery, traceable, stating the energies and the radionuclides, plus a written statement of what recalibration will cost and how long the instrument will be away.

Calibration, and the check source that keeps it honest

Calibration is annual, traceable, and specific to energy. An instrument calibrated against a caesium 137 beam is not thereby correct at 140 keV unless the certificate says what its response is there. Ask for the correction factors, and keep them with the instrument rather than in a file, because they are useless if the person holding the meter does not know they exist.

The practical problem in Bangladesh is turnaround. An instrument sent away for calibration is gone for weeks, and a department with one meter is a department that is uncalibrated or unmonitored for that period. There are two workable answers: hold two instruments and stagger their calibration dates, or write a loan instrument into the service agreement at purchase. Deciding this after the first calibration is due is how departments end up running for a year on an expired certificate.

Between calibrations, the daily check is what tells you the instrument still works. It takes under a minute: battery indication, background reading in a known clean place, then the response to the check source in a fixed geometry. Write all three in the log. The purpose is not the individual numbers, it is that a response which has fallen by a third since March tells you the tube is dying before the day you need it.

Using it so that it finds something

Most contamination surveys fail for the same reasons, and all of them are technique rather than equipment.

1. Take and record a background reading, in a clean area, every time. A survey without a background is a survey without a threshold.
2. Turn the audio on and watch the surface, not the meter.
3. Hold the probe close and steady, roughly a centimetre off the surface, and move it slowly. About one probe width per second is right. Sweeping quickly across a bench is the single most common way to walk past a spill.
4. Never touch the surface with the probe. A contaminated probe reports contamination everywhere for the rest of the day, and the mica window does not survive being wiped.
5. Survey in a fixed order, every time: bench, then the front of the L-block, then the floor in front of it, then the waste bin lid, then the door handle, then hands and shoes at the door. A route that is written down gets followed.
6. Do not survey while wearing the gloves you have just been working in. Change them first.
7. When something is found, mark it, record the reading and the background, and then decide whether it is fixed or removable by taking a wipe and counting it. A monitor alone cannot tell you which, and the answer changes what you do next.

Wipe counting needs a counter, not a survey meter. A well type counter is the proper instrument. Departments without one sometimes count wipes in the dose calibrator well, which is crude, geometry dependent and only defensible if the same method is used consistently and the limitation is written into the procedure.

The last point is about where the survey happens. Departments survey the hot lab bench thoroughly and forget the route the dose travels: the injection room, the arm of the chair, the trolley handle, the corridor floor between the two, and the toilet used by patients after injection. Contamination is found where people have walked, not where the source was kept. The waste side of the same problem is covered in radioactive waste in a nuclear medicine department, and what to do in the first hour after a spill is set out in the first hour after a spill in the hot lab.

Back to all Insights